2328 Phys. Chem. Chem. Phys., 2013, 15, 2328--2336 This journal is c the Owner Societies 2013 Cite this: Phys. Chem. Chem. Phys., 2013, 15, 2328 Temperature effects in dye-sensitized solar cells Sonia R. Raga and Francisco Fabregat-Santiago* In the standard solar cell technologies such as crystalline silicon and cadmium telluride, increments of temperature in the cell produce large variations in the energy conversion efficiency, which decreases at a constant rate. In dye solar cells the efficiency remains roughly constant with a maximum at around 30–40 1C and further decays above this temperature. In this work, the origin of this characteristic behavior is explained. Data show that under illumination recombination kinetics in the active layer of the cell is the same between 7 and 40 1C. Consequently, the efficiency of the cell remained virtually constant, with only small differences in the fill factor associated with changes in the series resistance. A further increase in temperature up to 70 1C produces an increase in recombination kinetics yielding lower photopotential and device performance. Finally, it is emphasized that at the normal operating temperatures of solar cells, the gap among the conversion efficiency of different technologies is much smaller than generally acknowledged. Introduction Dye sensitized solar cells (DSCs) are a promising alternative to silicon or thin film solar cells due to the low cost materials and simplicity in their fabrication process. These characteristics make DSC technology a candidate for large scale production of cheap energy. 1,2 The highest efficiencies reached with DSCs, 11–12%, 3–6 have been obtained using TiO 2 nanostructures for collecting the electrons, a redox electrolyte based on I /I 3 for transporting the holes and a ruthenium dye attached to the semiconductor nanoparticles for absorbing the radiation and separating the charges into the different charge carrier media. 7,8 A recent publication has also reported 12% efficiency with porphyrin dyes and cobalt based redox electrolyte. 9 Some of the beneficial characteristics of DSCs are that they maintain a high efficiency under diffuse or low light incident angle and that their performance presents little changes at environ- mental temperatures ranging between 20 and 80 1C. 10–13 Power conversion efficiency drops for rising temperatures in the case of the most extended technologies, at rates around 0.45% K 1 in mono- and polycrystalline silicon cells and of 0.25% K 1 for cadmium telluride. 14 In normal operation, cell temperatures (NOCT) rise to about 45–50 1C and consequently, their efficiency drops from nominal values approximately 10% for Si and 5% for CdTe. Dye solar cells instead present a maximum in the energy conversion efficiency close to NOCT. 12 This is a key aspect in the discussion of best performing solar cell technologies: as DSCs for 17 cm 2 modules have reached 9.9% efficiency, 15 the increase in efficiency at NOCT together with the decrease in the other technologies reduces the efficiency gap under the real operating conditions to a short distance. From a more fundamental view, the progress in the field of DSCs has been very exciting in the last few years from both points of view, the enhancement of the device performance and the development of the models needed to understand the origin of the behavior of DSCs. Very relevant contributions have been focused on describing the mechanisms that control parameters such as the short circuit photocurrent, J sc , open circuit photopotential, V oc , and fill factor, FF, that determine the performance of the solar device. 16–19 In general, charge collection is well resolved in DSCs, with efficiencies close to 100% in good cells. Photovoltage, however, presents large variations affected by charge recombination between TiO 2 , dye or electrolyte composition and also by the energetic position of the TiO 2 conduction band in relation to the redox energy level in the electrolyte. 20–22 Recombination of charge plays a major role in the photopotential attained by the DSC 17,22 and it is determined by different mechanisms that involve variations in acceptor species: Miyashita et al. 23 proposed that there is a locally increased concentration of I 3 near the dye under illumination and Boschloo and Hagfeldt 16 remarked the generation of other acceptor species as I 2 . On another hand O’Regan and Durrant 24 suggested that recombination increases Photovoltaic and Optoelectronic Devices Group, Departament de Fı ´sica, Universitat Jaume I, 12071 Castello´, Spain. E-mail: fran.fabregat@fca.uji.es; Fax: +34-964-729218; Tel: +34-964-387537 † Electronic supplementary information (ESI) available: Measures of performance of DSC for 500 h lifetime tests, Nyquist plots showing series resistances, values of conduction band shifts, plots of capacitance values vs. V ecb , comparison of capacitance in the dark and under illumination, comments on transport resis- tance and performance data of DSC after shifting the applied potential the displacement in the conduction band edge (DE c ). See DOI: 10.1039/c2cp43220j Received 13th September 2012, Accepted 10th December 2012 DOI: 10.1039/c2cp43220j www.rsc.org/pccp PCCP PAPER Published on 08 January 2013. Downloaded by OIST Graduate University on 05/12/2016 00:41:47. View Article Online View Journal | View Issue